Developmental trajectories of receptive and expressive communication in children and young adults with cerebral palsy

Transcription

1 DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY ORIGINAL ARTICLE Developmental trajectories of receptive and expressive communication in children and young adults with cerebral palsy RIMKE C VOS 1 ANNET J DALLMEIJER 1 MARJOLEIN VERHOEF 2,3 PETRA E M VAN SCHIE 1 JEANINE M VOORMAN 4 DIANA J H G WIEGERINK 5 JOKE J M GEYTENBEEK 1 MARIJ E ROEBROECK 5,6 JULES G BECHER 1 ON BEHALF OF THE PERRIN+ STUDY GROUP* 1 Department of Rehabilitation Medicine and EMGO+ Institute for Health and Care Research, VU University Medical Center, Amsterdam; 2 Brain Center Rudolf Magnus and Center of Excellence for Rehabilitation Medicine, University Medical Center and De Hoogstraat Rehabilitation, Utrecht; 3 Partner of NetChild, Network for Childhood Disability Research, Utrecht; 4 Rehabilitation Center De Trappenberg, Almere; 5 Transition and Lifespan Research Group, Department of Rehabilitation Medicine, Erasmus MC-University Medical Center, Rotterdam; 6 Rijndam Rehabilitation Center, Rotterdam, the Netherlands. Correspondence to Annet Dallmeijer, Department of Rehabilitation Medicine, VU University Medical Center, P.O. Box 7057, 1007 MB Amsterdam, the Netherlands. *Members of the group are listed in Appendix S1, available online. This article is commented on by Pennington on pages of this issue. PUBLICATION DATA Accepted for publication 7th March Published online 29th April ABBREVIATIONS BSCP Bilateral spastic cerebral palsy ICF-CY International Classification of Functioning, Disability and Health Children and Youth version NSCP Non-spastic cerebral palsy USCP Unilateral spastic cerebral palsy VABS Vineland Adaptive Behavior Scales AIM The aim of this study was to determine the developmental trajectories of expressive (speech) and receptive (spoken and written language) communication by type of motor disorder and in individuals with cerebral palsy (CP). METHOD The development of 418 participants (261 males, 157 females; mean age 9y 6mo [SD 6y 2mo], range 1 24y; Gross Motor Function Classification System (GMFCS) level I [n=206], II [n=57], III [n=59], IV [n=54], V [n=42]) was followed for 2 to 4 years in a longitudinal study. Communication performance was measured using the Vineland Adaptive Behavior Scales. The type of motor disorder was differentiated by type of CP as unilateral spastic (USCP, n=161), bilateral spastic (BSCP, n=202), and non-spastic (NSCP, n=55), while was determined by IQ or school type (regular or special). A multilevel analysis was then used to model the developmental trajectories. RESULTS The most favourable development of expressive communication was seen in USCP (vs BSCP b [SE]-2.74 [1.06], NSCP b [SE]-2.67 [1.44]). The difference between the development trajectory levels of children with and without was smaller for children with USCP than for those with BSCP and NSCP. For receptive communication, the most favourable development was found for all children with USCP and for BSCP or NSCP without (vs b [SE]-4.00 [1.16]). Development of written language was most favourable for children without (vs b [SE] [2.85]). INTERPRETATION The development of expressive communication was found to be most closely related to type of motor disorder, whereas the development of receptive communication was found to be most closely related to. Individuals with cerebral palsy (CP) may experience communication difficulties from infancy onwards. 1,2 The difficulty of these communication problems ranges from severe (non-speaking) to mild (the individual requires a slower pace of conversation as extra time is needed to understand or compose a message). 1 Communication is essential for social participation in society; 3,4 moreover, communication difficulties are associated with problems in child parent interaction and social participation with other significant persons. 4 In the International Classification of Functioning, Disability and Health Children and Youth version (ICF- CY), communication performance is one of nine domains covered by the ICF construct Activities & Participation. Communication performance is defined as a person transmitting and receiving messages in different environments: a person transmits a message by speech (expressive communication) and receives a message by means of spoken and written language (receptive communication). 1 The communication difficulties associated with CP are multifactorial 2 and are related not only to the characteristics of the condition (gross motor impairment, vision, hearing, epilepsy), but also to personal (education, behavioural problems) and environmental (siblings, parental stress, social economical status) factors. 5,6 These factors all potentially influence the communication performance of the individual. In previous studies, differences in communication difficulties have often been described in relation to 2014 Mac Keith Press DOI: /dmcn

2 the severity of CP according to their Gross Motor Function Classification System (GMFCS) level. 4,7 Using this classification system, a different (and less favourable) developmental trajectory of communication performance was found only for individuals functioning at GMFCS level V. 7 Differences in developmental trajectories between individuals functioning at GMFCS levels I to IV were less pronounced. 6 A large European study showed that differences in communication performance for individuals with CP is less related to their GMFCS level and more related to the type of motor disorder they have. 8 Motor disorders are classified into types based on abnormal patterns of posture and/or movement, in combination with measurements of muscle tone and limb involvement, including spastic CP (posture and movement dependent tone regulation disorder) with one (unilateral) or both (bilateral) sides of the body involved. Other motor disorders (ataxic CP and dyskinetic CP) are described as non-spastic CP. 9 The aforementioned European study found that individuals with non-spastic CP had greater communication difficulties compared with individuals with spastic CP. 8 In addition to factors related to CP, a strong association between communication difficulties and has also been reported. 5,7,8 Individuals with intellectual disability have more problems with expressive communication (speech) and receptive communication (comprehension of spoken and written language) than do those without. 5,7 To the best of our knowledge, no data are yet available on the developmental trajectories of expressive and receptive communication within types of motor disorder, or of the influence of on these trajectories. Communication is an essential aspect of social participation and inclusion in society is the targeted goal of rehabilitation treatment. Therefore, as communication is essential for interaction between persons in daily living and communication difficulties interfere with natural interactions, greater insight into the development of communication performance is needed. 2 The findings of the present longitudinal study may identify subgroups at risk of communication difficulties and may enhance the development of programmes to enable individuals with CP to participate more independently in social interactions. The aim of this study, therefore, was to determine the developmental trajectories of expressive (speech) and receptive (spoken and written language) communication by type of motor disorder and in individuals with CP in the Netherlands. METHOD Study design and participants This study was performed as part of the prospective longitudinal research programme PEdiatric Rehabilitation Research In the Netherlands (PERRIN), which was launched in 2000 as a collaboration between four What this paper adds This study indicates different developmental trajectories of expressive and receptive communication skills in cerebral palsy (CP). Expressive communication development is more closely related to the type of motor disorder. Expressive communication development was most favourable in unilateral spastic CP. The development of receptive communication skills is most closely related to intellectual (dis)ability. university medical centres and several rehabilitation centres in the Netherlands. For the present study, data on 418 children with CP were combined from four age cohort studies in the PERRIN research programme: 0 5 (aged 1 and 2y at baseline, n=93), 5 9 (aged 5 and 7y at baseline, n=114), 9 16 (aged 9, 11, and 13y at baseline, n=108), and (aged 16 20y at baseline, n=103). Within each age cohort, data were collected yearly over the course of 2 years (5 9) or 3 years (0 5, 9 16), or bi-yearly over 4 years (16 24; see Fig. 1). The recruitment process of these studies has been described in detail elsewhere In short, all eligible individuals had a clinical diagnosis of CP. Individuals were excluded if they had been diagnosed with additional diseases or disorders affecting motor functioning or if they or their caregiver lacked a basic knowledge of the Dutch language. Young adults with an (based on school type; regular or special) were also excluded. 10 The young adult cohort study aimed to determine the course and determinants of daily activities and participation from the perspective of the young adults themselves. The instruments used to assess this information were not suitable for individuals with and, therefore, they were excluded from the study. Informed consent was obtained from each patient and/or their parents or formal caregiver. Ethical approval for the study was obtained from the medical ethics committee of each participating centre. Characteristics of CP The condition of individuals with a clinical diagnosis of CP was described by functional mobility, classified using the Gross Motor Function Classification System (GMFCS) as levels I to V, and by type of motor disorder: unilateral spastic CP (USCP), bilateral spastic CP (BSCP), or nonspastic CP (NSCP, including dyskinetic, ataxic). The spread of type of motor disorder, epilepsy (more than one episode in the past year), hearing impairment, and vision impairment across the age cohorts is shown in Table I. Measures of communication performance Communication performance was determined using the validated Dutch version of the Vineland Adaptive Behavior Scales (VABS) survey. 14,15 The VABS survey is a reliable and validated instrument, 14,16,17 developed to assess the performance of children aged from 0 to 18 years 11 months, with or without any disability. 17 The VABS survey measures the child s performance by means of a semi-structured interview with the parents or caregivers 952 Developmental Medicine & Child Neurology 2014, 56:

3 770 eligible and invited to participate 341 non-participants (mostly language problems, participation in an other study or family stress) 429 consented to participate Five excluded (unforeseen additional disorders) Four excluded missing subtype Two exclude missing intellectual ability 418 included in analyses Complete data set, n=341 Missing 1 measurement, n=50 Missing 2 measurement, n= observations included in analyses Figure 1: Flow diagram of study population. according to current functioning of the child in daily living (irrespective of the use of aids). One of the four domains is communication, which contains 67 items on how a child uses expressive communication (e.g. uses phrases containing a noun and a verb, or two nouns ), receptive communication (e.g. follows instructions requiring an action and an object ), and uses and understands written language (e.g. prints or writes at least 10 words from memory ). Items are listed in developmental order and are divided into the following response categories: 0 (never performed), 1 (sometimes or partially performed) or 2 (usually or habitually performed). Accordingly, raw communication scores ranged between 0 and 134 and were compared with the VABS reference values, derived from 3000 children with typical development in north-west America. For the toddlers, the Dutch translation of the VABS screener was used, containing only the relevant items for this age group (expressive and receptive domain, not written language). 18 For the young adults, only data on written language were assessed. Intellectual disability The measurement for was determined through IQ testing either directly (Snijders Oomen nonverbal intelligent test [SON-R 19 ] for toddlers [n=93] and Raven s Coloured Progressive Matrices 20 for children [n=114]) or indirectly according to school type (adolescents [n=108] and young adults [n=103]). No was defined as IQ 70 (n=128) or attending a regular education programme (n=180) (in a regular school or in a school providing special education for physically disabled children). Intellectual disability was defined as an IQ <70 (n=79) or attending a special education programme in special schools (n=31) for children with an intellectual Trajectories of Communication in Children and Young Adults with CP Rimke C Vos et al. 953

4 Table I: Participant characteristics at baseline (n=418) Characteristic Unilateral spastic cerebral palsy Bilateral spastic cerebral palsy Non-spastic cerebral palsy Total (n) Without, n (%) Total 142 (88) 136 (67) 30 (55) 308 (74) Toddlers (0 5y) 30 (21) 18 (13) 1 (3) 49 (16) GMFCS level (I III/IV V), n 30/0 13/5 1/0 44/5 Epilepsy 7 (23) 4 (21) 0 (0) 13 (27) Hearing impairment 0 (0) 0 (0) 0 (0) 0 (0) Vision impairment 2 (7) 1 (6) 0 (0) 3 (6) Children (5 9y) 35 (25) 37 (27) 7 (24) 79 (26) GMFCS level (I III/IV V), n 35/0 33/4 7/0 75/4 Epilepsy 4 (11) 1 (3) 0 (0) 5 (6) Hearing impairment 1 (3) 0 (0) 1 (14) 2 (3) Vision impairment 8 (23) 13 (35) 1 (14) 22 (28) Adolescents (9 16y) 34 (24) 34 (25) 9 (30) 77 (25) GMFCS level (I III/IV V), n 34/0 25/9 7/2 66/11 Epilepsy 2 (6) 1 (3) 0 (0) 3 (4) Hearing impairment 2 (6) 1 (3) 0 (0) 3 (4) Vision impairment 2 (6) 4 (12) 2 (22) 8 (10) Young adults (16 24y) 43 (30) 47 (35) 13 (43) 103 (33) GMFCS level (I III/IV V), n 43/0 34/13 13/0 90/13 Epilepsy 4 (9) 1 (2) 0 (0) 5 (5) Hearing impairment 1 (2) 1 (2) 0 (0) 2 (2) Vision impairment 2 (5) 5 (11) 1 (8) 8 (8) disability, n (%) Total 19 (12) 66 (33) 25 (45) 110 (26) Toddlers (0 5y) 11 (58) 31 (47) 2 (8) 44 (40) GMFCS level (I III/IV V), n 9/2 11/20 0/2 20/24 Epilepsy 8 (73) 15 (48) 1 (50) 24 (55) Hearing impairment 1 (9) 3 (10) 1 (50) 5 (11) Vision impairment 4 (36) 8 (26) 0 (0) 12 (27) Children (5 9y) 5 (26) 19 (29) 11 (44) 35 (32) GMFCS level (I III/IV V), n 5/0 7/12 4/7 16/19 Epilepsy 1 (20) 8 (42) 4 (36) 13 (37) Hearing impairment 1 (20) 0 (0) 5 (45) 6 (17) Vision impairment 0 (0) 8 (42) 3 (27) 11 (31) Adolescents (9 16y) 3 (16) 16 (24) 12 (48) 31 (28) GMFCS (I III/IV V), n 3/0 5/11 3/9 11/20 Epilepsy 3 (100) 3 (19) 4 (33) 10 (32) Hearing impairment 1 (33) 1 (6) 1 (8) 3 (10) Vision impairment 1 (33) 6 (38) 7 (58) 14 (45) Total 161 (39) 202 (48) 55 (13) 418 (100) Values are expressed as n (%) except for Gross Motor Function Classification System (GMFCS) levels. disability based on IQ testing (with or without physical disabilities). Statistical analysis Descriptive statistics were computed for CP characteristics using SPSS software, version 20 (IBM SPSS Statistics; IBM Corp., Armonk, NY, USA). Multilevel analyses were performed using MLwiN (version 2.21; Graduate School of Education, University of Bristol, Bristol, UK) to analyse the developmental trajectories of the communication subdomains. Raw scores of domains of expressive communication, receptive communication, and written language were used. Because data for several age cohorts were combined, with repeated measures for the same patient, three levels were defined: observations (level 3) were clustered within participants (level 2) and participants were clustered within age cohorts (level 1). The developmental trajectories of the communication subdomains were modelled by type of motor disorder and for children and adolescents aged from 1 year to 16 years (expressive and receptive communication) or from 5 to 24 years (written language). Data on individuals with and USCP were available up to age 11 years and for those with BSCP and NSCP up to age 16 years. Based on the observed data, age was included in the model as an independent continuous variable, both linear (age) and squared (age 2 ). Type of motor disorder was included as two dummy variables (with USCP as reference variable) and as a dichotomous variable (with no as reference category). Each model of the communication subdomains included age and age 2. The likelihood ratio test was used to evaluate whether a random regression coefficient for age and age 2 needed to be considered in the models. Next, the type of motor disorder and were added to the model one by one to evaluate the influence on the communication subdomains. To evaluate the influence of the type of motor disorder and on the developmental trajectory of the communication subdomains, a two-way interaction of age (both linear and squared) by 954 Developmental Medicine & Child Neurology 2014, 56:

5 type of motor disorder and by was included, and the three-way interactions of age (both linear and squared) by type of motor disorder and by intellectual disability were entered into the model. The likelihood ratio test was used at each step to evaluate the additional value of the added determinant and the Wald statistic (for 1df) was used to evaluate the significance of the relationship between the outcome measures and determinants. 21 The reference category was alternated to determine differences between the developmental trajectories of the type of motor disorder. RESULTS The complete data set included 1312 observations of 418 participants ([261 males, 157 females; mean age 9y 6mo [SD 6y 2mo], range 1 24y; and by [with, n=110; Table II]). Overall, all GMFCS levels were represented in the study population: level I, n=206; level II, n=57; level III, n=59; level IV, n=54; and level V, n=42 (see Appendix S2, online supporting information). The regression coefficients of the final models of the developmental trajectories of communication performance on expressive communication, receptive communication and written language are shown in Table II. It was not necessary to include a random regression coefficient in any of the models (based on the ratio likelihood test). The developmental trajectories of the individuals with CP and those of children with typical development (derived from the 3000 children from north-west America) are included in the figures as a reference. Expressive communication The developmental trajectories (ages 1 16y) of expressive communication differed by type of motor disorder. The most favourable development trajectory (highest level) of expressive communication was found for children with USCP (p<0.01 vs BSCP and NSCP), followed by those with BSCP (p=0.01 vs NSCP). The difference between the trajectory levels of children with and without intellectual disability was smaller for children with USCP than for those with BSCP or NSCP (Fig. 2a). Receptive communication The developmental trajectories (age 1 16y) of receptive communication were not significantly different between types of motor disorder. However, the trajectories were dependent on type of motor disorder and different between children with and without (Table II, based on two-way interactions between age and Table II: Regression coefficients of the developmental trajectories of expressive communication, receptive communication, and written language by type of motor disorder Vineland Adaptive Behavior Scales communication subdomains Expressive communication Receptive communication Written language Type of motor disorder Regression coefficient (SE) Wald statistic (1df) Regression coefficient (SE) Wald statistic (1df) Regression coefficient (SE) Wald statistic (1df) Constant Age (0.58) 1.55 (0.22) 7.96 (0.68) Age (0.03) 0.06 (0.01) 0.22 (0.03) Unilateral spastic CP 0 ref 0 ref 0 ref Bilateral spastic CP 4.19 (3.42) ns 1.65 (1.16) ns 0.05 (2.28) ns Non-spastic CP 7.86 (6.98) ns 2.31 (2.49) ns 5.30 (3.40) ns Age 9 unilateral spastic CP 0 ref 0 ref 0 ref Age 9 bilateral spastic CP 2.43 (0.47) < (0.27) ns 0.12 (0.38) ns Age 9 non-spastic CP 5.79 (1.32) < (0.45) ns 0.41 (0.62) ns Age 2 9 unilateral spastic CP 0 ref 0 ref 0 ref Age 2 9 bilateral spastic CP 0.12 (0.04) < (0.02) ns 0.01 (0.01) ns Age 2 9 non-spastic CP 0.30 (0.06) < (0.02) ns 0.01 (0.02) ns Intellectual disability 0.56 (3.51) ns 4.00 (1,16) < (2.85) <0.01 Age (1.02) ns 0.79 (0.35) (1.05) <0.01 Age (0.07) ns 0.04 (0.03) ns 0.30 (0.09) <0.01 Age 9 unilateral spastic CP 9 0 ref 0 ref 0 ref Age 9 bilateral spastic CP (1.06) (0.36) (0.76) ns Age 9 non-spastic CP (1.44) ns 2.02 (0.49) < (0.08) <0.01 Age 2 9 unilateral spastic CP 9 0 ref 0 ref 0 ref Age 2 9 bilateral spastic CP (0.09) ns 0.03 (0.03) ns 0.15 (0.08) 0.05 Age 2 9 non-spastic CP (0.11) ns 0.13 (0.04) < (0.09) 0.04 ns, not significant. Trajectories of Communication in Children and Young Adults with CP Rimke C Vos et al. 955

6 a VABS expressive (raw score) Age (years) No USCP No BSCP No NSCP disability USCP disability BSCP disability NSCP Typically developing b VABS receptive (raw score) Age (years) disability USCP disability BSCP disability NSCP disability USCP disability BSCP disability NSCP Typically developing c VABS written language (raw score) Age (years) disability USCP disability BSCP disability NSCP disability USCP disability BSCP disability NSCP Typically developing Figure 2: Modelled developmental trajectories of (a) expressive communication and (b) receptive communication, for ages 1 to 16 years, and (c) written language, for ages 5 to 24 years, by type of motor disorder. The higher scores for unilateral spastic cerebral palsy (USCP) without are model driven. In addition, the delayed rate of development compared with the children with typical development is model driven. The decreased trajectory of receptive communication of bilateral spastic cerebral palsy (BSCP) is the result of only a few data points after the age of 12 years. No data were available for children older than 11 years with USCP and. NSCP, non-spastic cerebral palsy; VABS, Vineland Adaptive Behavior Scales. ). In children with USCP, no differences in trajectory levels were found between children with and without, while in children with BSCP or NSCP, lower and less favourable trajectories were found for children with (Fig. 2b). 956 Developmental Medicine & Child Neurology 2014, 56:

7 Written language The developmental trajectories (age 5 24y) of the use and understanding of written language were not significantly different between types of motor disorder. However, as with receptive communication, trajectories differed for children with and without (Table II, based on the two-way interaction between age and type of motor disorder) and dependent on the type of motor disorder. Development started at the age of 5 years and showed a steep incline among children without, with a smaller difference in children with and USCP than in those with BSCP or NSCP (Fig. 2c). DISCUSSION In the present study, developmental trajectories for communication performance (expressive communication, receptive communication, and written language) in children and adolescents with CP in the Netherlands were estimated by type of motor disorder and. Type of motor disorder and were both related to communication performance. More importantly, these relationships differed for expressive communication (speech), receptive communication (spoken and written language). Type of motor disorder was most closely related to the developmental trajectory of expressive communication, whereas was most closely related to the developmental trajectories of receptive communication. Examining the influence of type of motor disorder and in more detail and in relation to the different subdomains of communication, we found that expressive communication (speech) was most affected in BSCP and NSCP, in agreement with other researchers. 22 This relationship is to be expected, as speech is especially sensitive to and affected by involuntary face muscle movements and loss of voluntary motor control, both of which are more often present in children with BSCP and NSCP. Our finding that children with have greater difficulties with expressive communication than those without also agrees with earlier reports. 7,22 However, in the present study this finding was dependent on the type of motor disorder, something that was not reported in previous studies. Instead, earlier studies focussed on the relationship between expressive communication and in relation to severity of CP (GMFCS 22 or a three-level Gross Motor Scale 7 ) with conflicting findings. Thus, the present results emphasize the value of examining communication difficulties in relation to the additional information provided by type of motor disorder. Receptive communication (comprehension of spoken language and written language) was influenced more by than by type of motor disorder. Children and adolescents without showed fewer difficulties with receptive communication. It was already known from previous cross-sectional studies that children without have fewer difficulties with language and the use of sentences. 5,7,22 The relevance of the current findings is that we were able to determine the influence of on the development trajectory of receptive communication skills. The substantial influence of on receptive communications skills (processing and understanding spoken and written language) reflects the complexity of this communication subdomain and emphasizes the relationship between verbal intelligence and receptive communication skills. 23 However, the receptive communication skills mentioned above provide only a general indication of a child s intellectual skills. In the present study the measurement for was based both on direct and indirect (based on school type) IQ testing. Consequently, it was not possible to differentiate between mild (IQ=50 70) and severe (IQ<50). Moreover, for the children with the most severe levels of lowered gross motor functioning, other instruments are more appropriate (e.g. Computer-Based instrument for Low motor Language Testing (C-BiLLT) for measuring the child s level of communication skills, and should be used to tailor teaching approaches. 24 It is unclear whether the less favourable development of receptive communication due to is influenced by educational factors (school type). Based on test results before entering, special education is more tailored to the child s needs. In order to prevent under- or overestimation of the child s level of communication skills, accurate identification of the child s level of communication skills is vital in ensuring that the most appropriate educational targets and teaching strategies are used. For effective communication in daily living, it is necessary to both express (send) messages as well as to receive and understand messages, and to be able to alternate between the two. Therefore, analogous to the GMFCS, a classification system for communication in children with CP was recently developed; the Communication Function Classification System. 25 A limitation of the current study is that the Communication Function Classification System was not included in the data set because it was not yet developed at the time of measurement. The VABS measures expressive and receptive communication from the caregiver s perspective, which is potentially open to bias, especially for those patients with severe motor impairment. Furthermore, we classified the type of motor disorder according to the Surveillance of Cerebral Palsy in Europe. According to this classification, the BSCP subtype is the most heterogeneous and includes children with a predominant involvement of the legs (diplegia), as well as children in whom both arms and legs are more or less severely affected (quadriplegia). It is most likely that children with a predominantly leg involvement will experience fewer difficulties in expressive speech; however, this was not examined in the present study. In addition, it is most likely that for toddlers the dyskinetic subtype will not yet have developed. The results of the BSCP subtype as a group Trajectories of Communication in Children and Young Adults with CP Rimke C Vos et al. 957

8 could, therefore, be more favourable than for children with quadriplegic involvement. In addition, the presence of epilepsy and vision and/or hearing impairments can influence the communication performance of the child. Although the incidence of these disease characteristics is available and presented for each age cohort, the number of observations in the present study did not allow further analysis or evaluation into the influence of these characteristics on the developmental trajectories. Despite this, the present study makes a noteworthy contribution to the field of rehabilitation medicine as a result of the longitudinal study design used. Longitudinal data are essential when studying clinical change in individuals with CP, something that is not possible with cross-sectional studies. Future research should focus on the developmental trajectories of expressive and receptive communication of non-verbal children with CP. In addition, for both verbal and non-verbal children future research with larger sample sizes is needed to determine the potential influence of other disease characteristics (i.e. epilepsy, vision and hearing impairments, or personal and environmental factors) on these developmental trajectories. CONCLUSION Knowledge of the developmental trajectories of expressive and receptive communication skills is useful. The results of the present study show that development of expressive communication, although based on parent report and, therefore, open to bias, is more closely related to the type of motor disorder. In addition, receptive communication is more closely related to the intellectual ability to process and understand others and to understand the meaning of written language. This emphasizes the importance of accurately assessing the receptive communication skills of children with CP. ACKNOWLEDGEMENTS This research was undertaken as part of the PEdiatric Rehabilitation Research In the Netherlands (PERRIN) research programme and has been supported by the Stichting Rotterdams Kinderrevalidatie Fonds Adriaanstichting and Stichting Johanna KinderFonds (2010/0040). The authors thank the children and young adults who participated in this study, the university medical centres and rehabilitation centres for providing the data, Professor JWR Twisk (VU University Medical Center, Amsterdam) for his statistical support, and especially commemorate Professor E Lindeman (University Medical Centre Utrecht), who passed away in September 2012, and who was very much involved in the conceptualization and design of the study programme. We also thank all members of the PERRIN + Study Group. The authors have stated that they had no interests that might be perceived as posing a conflict or bias. SUPPORTING INFORMATION The following additional material may be found online: Appendix S1: Members of the PERRIN+ Study Groups. Appendix S2: Baseline characteristics of the participants. REFERENCES 1. World Health Organization. International Classification of Functioning, Disability and Health, Children and Youth Version (ICF). Geneva: WHO, Pennington L, Goldbart J, Marshall J. Speech and language therapy to improve the communication skills of children with cerebral palsy. Cochrane Database Syst Rev 2004; 2: CD Hidecker MJ, Ho NT, Dodge N, et al. Inter-relationships of functional status in cerebral palsy: analysing gross motor function, manual ability, and communication function classification systems in children. Dev Med Child Neurol 2012; 54: Pennington L, McConachie H. Predicting patterns of interaction between children with cerebral palsy and their mothers. Dev Med Child Neurol 2001; 43: Sigurdardottir S, Vik T. Speech, expressive language, and verbal cognition of preschool children with cerebral palsy in Iceland. Dev Med Child Neurol 2011; 53: Voorman JM, Dallmeijer AJ, van EM, Schuengel C, Becher JG. Social functioning and communication in children with cerebral palsy: association with disease characteristics and personal and environmental factors. Dev Med Child Neurol 2010; 52: Pirila S, van der Meere J, Pentikainen T, et al. Language and motor speech skills in children with cerebral palsy. J Commun Disord 2007; 40: Fauconnier J, Dickinson HO, Beckung E, et al. Participation in life situations of 8 12 year old children with cerebral palsy: cross sectional European study. BMJ 2009; 338: b Surveillance of Cerebral Palsy in Europe. (SCPE). Surveillance of cerebral palsy in Europe: a collaboration of cerebral palsy surveys and registers. Dev Med Child Neurol 2000; 42: Donkervoort M, Roebroeck M, Wiegerink D, Heijden- Maessen H, Stam H. Determinants of functioning of adolescents and young adults with cerebral palsy. Disabil Rehabil 2007; 29: Gorter JW, Ketelaar M, Rosenbaum P, Helders PJ, Palisano R. Use of the GMFCS in infants with CP: the need for reclassification at age 2 years or older. Dev Med Child Neurol 2009; 51: Smits DW, Gorter JW, Ketelaar M, et al. Relationship between gross motor capacity and daily-life mobility in children with cerebral palsy. Dev Med Child Neurol 2010; 52: e Voorman JM, Dallmeijer AJ, Knol DL, Lankhorst GJ, Becher JG. Prospective longitudinal study of gross motor function in children with cerebral palsy. Arch Phys Med Rehabil 2007; 88: de Bildt A, Kraijer D, Sytema S, Minderaa R. The psychometric properties of the Vineland Adaptive Behavior Scales in children and adolescents with mental retardation. J Autism Dev Disord 2005; 35: de Bildt AA, Kraijer DW. Vineland-Z, Dutch Version of the Vineland Adaptive Behavior Scales: Survey Forms Manual. Leiden: PITS, Rosenbaum P, Saigal S, Szatmari P, Hoult L. Vineland Adaptive Behavior Scales as a summary of functional outcome of extremely low-birthweight children. Dev Med Child Neurol 1995; 37: Sparrow SS, Balla D, Cicchetti DV. The Vineland Adaptive Behavior Scales (Interview edition, Survey Form). Circle Pines, MN: American Guidance Service Inc., van Duijn G, Dijkxhoorn Y, Noens I, Scholte E, van Berckelaer-Onnes I. Vineland Screener 0 12 years research version (NL). Constructing a screening instrument to assess adaptive behaviour. Int J Methods Psychiatr Res 2009; 18: Tellegen P, Winkel M, Wijnberg-Williams B, Laros J. Snijders-Oomen Niet Verbale Intelligentietest SON-R 1,5-7. Handleiding en Verantwoording. Lisse: Swets & Zeitlinger, Raven J, Raven JC, Court JH. Section 2: the coloured progressive matrices. Manual for Raven s Progressive Matrices and Vocabulary Scales. San Antonio, TX: Harcourt Assessment, Developmental Medicine & Child Neurology 2014, 56:

9 21. Twisk JWR. Applied Multilevel Analysis; A Practical Guide for Medical Researchers. Cambridge: Cambridge University Press, Parkes J, Hill N, Platt MJ, Donnelly C. Oromotor dysfunction and communication impairments in children with cerebral palsy: a register study. Dev Med Child Neurol 2010; 52: Clark EV. How language acquisition builds on cognitive development. Trends Cogn Sci 2004; 8: Geytenbeek JJ, Mokking LW, Knol DL, Vermeulen RJ, Oostrom KJ. Reliability and validity of the C-BiLLT: a new instrument to assess comprehension of spoken language in young children with CP and complex communication needs. Augmentative Altern Commun 2014; (in press). 25. Hidecker MJ, Paneth N, Rosenbaum PL, et al. Developing and validating the Communication Function Classification System for individuals with cerebral palsy. Dev Med Child Neurol 2011; 53: Trajectories of Communication in Children and Young Adults with CP Rimke C Vos et al. 959